黄河上游梯级水库群若干生态环境风险的分析方法与模型

With the background of the Yellow River upper reaches cascade development, this issue aims to conduct a study on risk identification and risk mechanism analysis method. Based on scientifically identification and cumulative effects analysis on cascade development, particularly on water pollution, eco-environmental water requirements deficit, sand sedimentation and other outstanding risk issues, reasonable method of risk analysis、uncertainty modeling solution and risk management methods would be researched . Improved genetic algorithm is adopted to optimize the solution of cascade reservoir group water quality grey-random composite uncertainty model, in order to overcome the drawback of single uncertainty analysis; based on risk factor hierarchy analysis method ,the fuzzy neural network model of eco-environment water requirements would be established. By reasonable determining risk factor weights , the prediction accuracy of eco-environment water requirement would be improved; to build GA-PLSR model of the sedimentation of the Cascades Group, effectively overcome the effects of multiple correlation between risk factors on the accuracy of models; and combined with the upper reaches of the Yellow River cascade reservoirs ,application study would be conducted, to put forward the risk evaluation and prevention measures .The research results could provide a scientific basis for risk management and decision-making of the basin cascade group eco-environment, and has important significance on promoting the cascade development of river basin and the eco-environment sustainable development.

以黄河上游水库梯级开发为背景，针对其影响显著的生态环境风险问题进行机理分析与方法研究。在对生态环境风险进行系统辨识和累积效应分析基础上，重点针对水质污染、生态环境需水量短缺、泥沙淤积等突出风险问题，研究合理的风险分析、不确定性建模求解和风险管理方法。采用改进遗传算法优化求解所建的梯级库群水质风险灰色-随机复合不确定性模型，克服单一不确定性分析的不足；建立基于风险因子层次分析法的生态环境需水量模糊神经网络模型，合理确定风险因子权重，改进生态环境需水量预测精度；基于泥沙风险分析，建立梯级库群泥沙淤积的GA-PLSR模型，有效克服风险因子间多重相关性对模型精度的影响；并结合黄河上游梯级水库群开展应用研究，提出风险分析评价与防范减缓措施。本项目的研究成果可为黄河及其它流域梯级库群的生态环境风险管理与决策提供科学依据，对于推进流域梯级开发与生态环境可持续发展具有重要意义。

流域梯级开发在带来巨大综合效益的同时，也造成了一系列生态环境问题。对流域梯级开发可能导致的生态环境影响及相关风险问题进行研究，是当今国际上水资源开发利用和生态环境保护中亟待解决的重要研究课题。.本项目以黄河上游水库梯级开发为背景，针对其影响显著的生态环境风险问题进行了机理分析与方法研究。在对生态环境风险进行系统辨识和累积效应分析基础上，重点针对水质污染、生态环境需水量短缺、泥沙淤积等突出风险问题，研究合理的风险分析、不确定性建模求解和风险管理方法。采用改进遗传算法优化求解所建的梯级库群水质风险灰色-随机复合不确定性模型，克服了单一不确定性分析的不足；建立了基于风险因子层次分析法的生态环境需水量模糊神经网络模型，合理确定风险因子权重，改进了生态环境需水量预测精度；基于泥沙风险分析，建立了梯级库群泥沙淤积的GA-PLSR模型，有效克服了风险因子间多重相关性对模型精度的影响；并结合黄河上游梯级水库群开展应用研究，提出了风险分析评价与防范减缓措施。.本项目的研究成果可为黄河及其它流域梯级库群的生态环境风险管理与决策提供科学依据，对于推进流域梯级开发与生态环境可持续发展具有重要意义。